forked frompostgres/postgres
- Notifications
You must be signed in to change notification settings - Fork6
Commit4576a69
committed
Fix yet another race condition in recovery/t/001_stream_rep.pl.
In commit5c77690, we added polling in front of most of theget_slot_xmins calls in 001_stream_rep.pl, but today's results frombuildfarm member nightjar show that at least one more poll loopis needed.Proactively add a poll loop before the next-to-last get_slot_xmins callas well. It may be that there is no race condition there because thestandby_2 server is shut down at that point, but I'm quite tired offighting with this test script. The empirical evidence that it's safe,from the buildfarm, is no stronger than the evidence for the othercall that nightjar just proved unsafe.The only remaining get_slot_xmins calls without wait_slot_xminsprotection are the first two, which should be OK since nothing hashappened at that point. It's tempting to ignore that special caseand merge get_slot_xmins and wait_slot_xmins into a single function.I didn't go that far though.Discussion:https://postgr.es/m/18436.1502228036@sss.pgh.pa.us1 parentb2c95a3 commit4576a69
1 file changed
+7
-1
lines changedLines changed: 7 additions & 1 deletion
Original file line number | Diff line number | Diff line change | |
---|---|---|---|
| |||
220 | 220 |
| |
221 | 221 |
| |
222 | 222 |
| |
| 223 | + | |
| 224 | + | |
| 225 | + | |
223 | 226 |
| |
224 | 227 |
| |
225 |
| - | |
| 228 | + | |
| 229 | + | |
226 | 230 |
| |
227 | 231 |
| |
228 | 232 |
| |
| |||
297 | 301 |
| |
298 | 302 |
| |
299 | 303 |
| |
| 304 | + | |
| 305 | + | |
300 | 306 |
| |
301 | 307 |
| |
302 | 308 |
| |
|
0 commit comments
Comments
(0)